The Condition of Forests

in Europe

Federal Research Centre for Forestry

and Forest Products (BFH)

Convention on Long-range Transboundary

Air Pollution: International Co-operative

Programme on Assessment and Monitoring

of Air Pollution Eff ects on Forests

European Union Scheme on the Protection

of Forests against Atmospheric Pollution

2002 Executive Report

United Nations

Economic Commission

for Europe

European Commission

© UNECE and EC, Geneva and Brussels, 2002

Reproduction is authorized, except for commercial purposes,

provided the source is acknowledged

Printed in Germany

The Condition of Forests

in Europe

2002 Executive Report

Convention on Long-range Transboundary

Air Pollution: International Co-operative

Programme on Assessment and Monitoring

of Air Pollution Eff ects on Forests

European Union Scheme on the Protection

of Forests against Atmospheric Pollution

United Nations

Economic Commission for Europe

European Commission

Federal Research Centre for Forestry

and Forest Products (BFH)

ISSN 1020-587X

Preface 6

Impacts of environmental stress factors on European forests

– overview and summary – 8

1. Introduction 10

2. Crown condition in 2001 and past developments 12

2.1 Introduction 12

2.2 Crown condition in 2001 and trends 12

2.3 Infl uences on crown condition 14

Th e infl uence of storms on forest ecosystems 17

Th e condition of the holm oak (Quercus ilex) 18

3. Present deposition and critical loads of nitrogen, acidity and metals for

forest ecosystems 20

3.1 Introduction 20

3.2 Nitrogen 20

3.3 Acidity 24

3.4 Heavy metals 24

3.5 Outlook 24

4. Ground vegetation and forest biodiversity 26

4.1 Introduction 26

4.2 Plant diversity at plot level 26

Biodiversity in European forests 27

International processes and projects 27

4.3 Single species in relation to environmental factors 28

4.4 Contribution of the monitoring programme to forest biodiversity

assessments - a future focus 28

5. Conclusions 30

Annex 32

C

Acknowledgements

Th e United Nations Economic Commission for Europe and the European Commission wish to express their appreci-

ation to all persons and institutions having contributed to the preparation of the report:

in particular the Federal Research Centre for Forestry and Forest Products - Programme Co-ordinating Centre of ICP

Forests, as well as the National Focal Centres for the data supply;

as well as to

R. Fischer (ed., ch. 1+2+5), W. de Vries (ch. 1+3+4), M. Barros (ch. 2 – Holm oak focus), H.van Dobben (ch. 4),

M. Dobbertin (ch. 2 – storm focus), H-D. Gregor (ch. 3 – critical loads focus), T-B. Larsson (ch. 4 – biodiversity focus),

M. Lorenz (summary), V. Mues (ch. 2), H.D. Nagel (ch. 3 – critical loads focus), P. Neville (ch. 4), G. Sanchez-Pena

(ch. 2 – Holm oak focus), D. de Zwart (ch. 4).

Th e designations employed and the presentation of material in this report

do not imply the expression of any opinion whatsoever

on the part of the Secretariat of the United Nations concerning the legal status

of any country, territory, city or area or of its authorities,

or concerning the delimitation of its frontiers or boundaries.

Th e views expressed in this report are the author’s and do

not necessarily correspond with those of the European Commission.

After approval by the Task Force of ICP Forests this report was derestricted by the Working Group on Eff ects

of the Convention on Long-range Transboundary Air Pollution

P

It is a great privilege and pleasure for

me to contribute with this preface to

the 2002 Executive Report on Forest

Condition in Europe. As a representa-

tive of the research community I strong-

ly believe that sustainability can only

be guaranteed with the use of sound

scientifi c information when develop-

ing public policies.

Th e forest monitoring programme of

EU and ICP Forests has become an es-

sential source of information for the

formulation of clean air policy. Th e

Programme has eff ectively increased

awareness in the scientifi c, political

and public areas of the eff ects of the

depositions of atmospheric pollutants

on forests.

Th is Report presents for the fi rst time

a Europe-wide assessment of critical

deposition loads based on measure-

ments from monitoring plots. Th ese

results are an important cornerstone

in the production of areal maps for

Europe.

Another novel feature of the Report is

the assessment of species diversity. It

shows clearly that soil acidity has a neg-

ative infl uence on ground vegetation

diversity in the forests. Th e inclusion

of the measurement of species diversi-

ty in the ICP Forests is also important

because species diversity serves as an

indicator of biological diversity in for-

est ecosystems e.g. for the Ministeri-

al Conference of Protection of Forests

in Europe (MCPFE).

Th e ICP Forests is very much orient-

ed towards empirical natural sciences.

Th is is understandable and acceptable

as a starting point. However, the orig-

inal reason for forest condition prob-

lems is usually to be found in econom-

ic development and human behaviour.

Consequently, solutions to the problems

are almost always connected with so-

cio-economic and political factors. Th is

suggests that the ICP Forests should

also include socio-economic and pol-

icy research in its agenda.

Although changes in the condition of

forests in Europe can only be traced

by means of continuous monitoring

and the collection of data, it is timely

to ask whether suffi cient attention has

been paid to the effi cient use of exist-

ing information and knowledge. Col-

lecting more and more data often be-

gins to dominate activities to the extent

that there are not enough time and re-

sources left for analyzing all informa-

tion available. In addition, policy mak-

ers and other users of research results

do not always consider the results and

conclusions produced by the research

community in their deliberations. In

many cases a major step forward would

be made if that what we already know

were implemented properly.

Despite the warning note above, I do

not suggest that the ICP Forests should

stop collecting new data. We need con-

tinuous monitoring, e.g. to see how the

implementation of clean air policies has

succeeded. My understanding is that

scientists and organizations also out-

side the Programme have extensively

utilized collected data, and many results

have been effi ciently adopted in polit-

ical decision-making. However, as an

external observer, I cannot help asking

whether it is still necessary to collect

so many new data. Shouldn’t more em-

phasis, at least temporarily, be placed

on compiling, analyzing and synthe-

sizing existing information at the ex-

pense of getting new fi gures?

Finally, I would like to congratulate the

ICP Forests and the EC on this excel-

lent work and off er the collaboration

of IUFRO - the International Union of

Forest Research Organizations. In addi-

tion to its almost 300 discipline-based

Research Groups and Working Parties,

IUFRO has established problem-orient-

ed Task Forces that cover many of the

key issues of the ICP Forests. In fact,

several scientists who currently work

in the Programme are also active with-

in the IUFRO framework.

Risto Seppälä

President

International Union of Forest Research

Organizations

F :

://..

Timberline in the mountains of northern Finland.

I

E

- –

� Th e crown condition of the forests

in Europe deteriorated considerably

during the fi rst decade of monitoring.

After some recuperation in the mid-

1990s the deterioration resumed. In

2001 more than 20 % of the sample

trees were rated as damaged.

� Th e defoliation of Scots pine and

beech is mainly related to the ef-

fects of weather extremes, in par-

ticular precipitation, age, insects,

fungi, and air pollution. Defoliation

of these species is correlated with

atmospheric inputs of sulphur.

� Depositions of nitrogen, acidi-

ty and heavy metals exceed criti-

cal loads over a large proportion of

the Intensive Monitoring Plots, in-

dicating enhanced risks for leach-

ing of aluminium and base cations,

tree root damage, crown damage by

drought, frost and pests and changes

in the plant diversity of ground veg-

etation.

� Plant diversity is mainly determined

by actual soil condition, specifi cal-

ly acidity, as well as nitrogen dep-

osition, temperature, precipitation

and tree species.

� Th e monitoring programme can

substantially contribute to environ-

mental policies in the context of bio-

diversity and sustainable forest man-

agement.

Th e United Nations Economic Commis-

sion for Europe (UNECE) and the Euro-

pean Union (EU) have been monitoring

forest condition in Europe jointly for 16

years. Th e above outlines main results

and conclusions of the programme as

presented in this annual report.

Th e monitoring system

Large-scale variations of forest condi-

tion over space and time are assessed in

relation to natural and anthropogenic

factors for more than 6 000 monitoring

plots spread over a systematic 16 x 16

km grid throughout Europe (Level I).

Detailed causal relationships are stud-

ied on 860 Intensive Monitoring Plots

covering the most important forest ec-

osystems in Europe (Level II). Changes

to the condition of forests in Europe,

their reaction to environmental stress

and the eff ects of environmental pol-

icies can only be traced by means of

continuous systematic and intensive

monitoring. With its large number of

plots and parameters and the partici-

pation of 39 countries, the programme

operates one of the world‘s largest bio-

monitoring networks.

Crown condition

Crown condition is a visible reaction

on numerous environmental factors af-

fecting tree vitality. Changes in crown

condition are therefore assessed annu-

ally for both Level I and Level II plots.

Th is provides in addition an indica-

tion of the forest health criterion for

the Ministerial Conference on the Pro-

tection of Forests in Europe (MCPFE).

Fifteen years of large-scale monitoring

of crown condition has revealed that

after initial deterioration followed by

some recuperation in the mid-1990s,

the deterioration has resumed.

Th e results of statistical evaluations de-

scribed in the present report confi rm

earlier fi ndings explaining the varia-

tion of defoliation mainly in terms of

the eff ects of weather extremes, in par-

ticular precipitation, insects, fungi, and

age. Also, relationships between defoli-

ation of Scots pine and beech and sul-

phur deposition are substantiated by

the recent statistical evaluations of the

transnational data set.

Atmospheric deposition

Of the various environmental factors

infl uencing forest condition, the pro-

gramme pays particular attention to

air pollution, in accordance with its

objectives. Sulphur depositions have

been reduced on many plots. Th is is a

clear success of the drastic reductions

of sulphur emissions in Europe under

the Convention on Long-range Trans-

boundary Air Pollution (CLRTAP) and

other pollution abatement strategies.

With regard to air pollution control,

a factor of particular concern is how

much pollution a particular environ-

ment can tolerate in the long term (crit-

ical loads). For the fi rst time this report

presents a European-wide assessment

of critical loads of forest soils based on

measurements from the Intensive Mon-

itoring Plots. Critical loads for nitrogen

accumulation in the soil were exceed-

ed at 92% of the plots investigated. In

earlier reports it was shown that nitro-

gen accumulations in the soil are the

predominant source of potential soil

acidifi cation. From saturated ecosys-

tems they can be released again, now

increasing nitrate concentrations in the

ground water. Calculations of critical

loads related to nitrogen concentra-

tions in the foliage of trees show that on

45% of the co nifer plots there is an in-

creased risk of drought stress, frost and

pests, whereas risks for ground vegeta-

tion changes occur on more than half

of the evaluated plots. Critical loads

for acidity that are related to eff ects

on tree roots were exceeded at 33% of

the investigated plots. Base cations or

aluminium pools in the soil might be

declining on up to 64% of the Inten-

sive Monitoring Plots. Critical loads

for lead and cadmium related to pos-

sible impacts on soil organisms were

exceeded on 92% and 29% of the plots,

respectively. It has to be taken into ac-

count that the presented critical loads

refer to a steady state. Th erefore the

time before eff ects become visible can

take several years to decades.

Ground vegetation

Th e plant diversity has been related to

environmental factors on nearly 200

Level II plots. Th e results show large dif-

ferences in plant diversity on the mon-

itored plots. Th e present occurrence of

species is mainly determined by actu-

al soil condition, specifi cally acidity,

temperature, precipitation, and tree

species. In the statistical models also

nitrogen deposition had a signifi cant in-

fl uence on ground vegetation. Th e pro-

gramme has recognised the importance

of the biodiversity issues and a newly

established working group is now re-

sponsible for intensifi ed assessments

and evaluations that might in the fu-

ture make it possible to quantify en-

vironmental impacts on fl oristic bio-

diversity in forests.

Future directions

Having set out to assess the eff ects of

air pollution on forests, the programme

has provided important information for

the implementation of clean air policies

under the UNECE and EU and will con-

tinue to do so in the future. However,

its well-established infrastructure, its

multidisciplinary monitoring approach

and its comprehensive database mean

that it also has signifi cant contributions

to make in other areas of internation-

al environmental politics.

It pursues the objectives of several of

the resolutions of the MCPFE and pro-

vides information on some of the quan-

titative indicators for sustainable forest

management. In addition, the soil data

of the programme is expected to be im-

portant for the assessment of carbon

sinks in the frame of the Kyoto Proto-

col under the Framework Convention

on Climate Change (FCCC). Contribu-

tions of the programme to the United

Nations Forum on Forests as well as

to the implementation of the Conven-

tion on Biological Diversity, including

the expanded work programme on for-

est biological diversity, are expected.

In addition, the programme is receiv-

ing increasing attention from research

institutions and policy-making bodies

outside Europe. An example is its re-

cently initiated co-operation with the

Acid Deposition Monitoring Network

in East Asia (EANET) and the close co-

operation with forest monitoring pro-

grammes in Canada and the USA.

F :

http://www.icp-forests.org

(ICP Forests)

http://europa.eu.int/comm/agriculture

(European Commission)

http://www.fi mci.nl

(Forest Intensive Monitoring Co-ordinating

Institute)

Mountain forest, Italy .

. I

Th e forest monitoring programme

in an international context

In the early 1980s a severe deteriora-

tion of forest condition was observed

in large areas of Europe. As a response

to growing concern about the role of air

pollution in this decline, the Interna-

tional Co-operative Programme on the

Assessment and Monitoring of Air Pol-

lution Eff ects on Forests (ICP Forests)

was established in 1985 under the UN-

ECE Convention on Long-range Trans-

boundary Air Pollution (CLRTAP).

In 1986 the European Union (EU)

adopted the Scheme on the Protec-

tion of Forests against Atmospheric

Pollution and with the Council Regu-

lation (EEC) No. 3528/86 the legal ba-

sis was provided for the co-fi nancing of

the assessments. Since then, ICP For-

ests and the EU have been co-operat-

ing closely in monitoring the eff ects of

air pollution and other stress factors on

forests. Th e activities pursue the ob-

jectives of resolutions Ministerial Con-

ference on the Protection of Forests in

Europe (resolution S1 - Strasbourg, H1

- Helsinki, L2 - Lisbon).

Today 39 countries are participating

in the monitoring programme, which

contributes to the implementation of

clean air policies under UNECE and

EU as well as at national levels.

Programme objectives

Th e objectives of the monitoring pro-

gramme are:

� to provide a periodic overview on the

spatial and temporal variation in for-

est condition in relation to anthro-

pogenic and natural stress factors for

a European and national large-scale

systematic network (Level I);

� to contribute to a better understand-

ing of the relationships between the

condition of forest ecosystems and

stress factors, in particular air pol-

lution, through intensive monitor-

ing of a number of selected perma-

nent observation plots spread over

Europe (Level II);

� to contribute to the calculation of

critical levels, critical loads and their

exceedances in forests;

� to collaborate with other environ-

mental monitoring programmes

in order to provide information on

other important issues, such as cli-

mate change and biodiversity in for-

ests and thus contribute to the sus-

tainable management of European

forests;

� to compile information on forest eco-

system processes and to provide pol-

icy makers and the public with rele-

vant information.

Monitoring arrangements

Th e objectives of the programme are

implemented by a systematic large scale

monitoring network (Level I) and an In-

tensive Forest Monitoring Programme

(Level II) (see Tab. 1-1).

At Level I approximately 6 000 perma-

nent plots are systematically arranged

in a 16 x 16 km grid throughout Eu-

rope. At these sites crown condition

is assessed annually. In addition, soil

and/or foliage surveys were conducted

on most of the plots. A new soil sur-

vey is under discussion.

For intensive monitoring, more than

860 Level II plots have been selected

in the most important forest ecosys-

tems of the countries participating. A

larger number of key factors are meas-

ured on these plots; the data collect-

ed can be used for case studies of the

more common combinations of tree

species and sites.

Key factors measured at both levels

form the basis for an extrapolation of

results. Th e inclusion of further pa-

rameters and surveys is currently be-

ing considered.

Th is year’s report

Th is year’s report provides the results

for the annual crown condition sur-

vey as well as in-depth evaluations

for specifi c tree species. Th e focus on

holm oak is the continuation of a se-

ries that has dealt with European and

sessile oak, common beech and Alep-

po pine in previous years. (Chapter 2).

As far as intensive monitoring is con-

cerned, the focus in last year’s report

was on the eff ect of sulphur and nitro-

gen in the forest ecosystem and the re-

sponse of the soils. Chapter 3 of this

year’s report focuses on critical depo-

sition loads in comparison to present

loads, whereas for next year’s report it

is planned to consider impacts under

various emission scenarios on forest

ecosystem condition in the future.

Since the objectives of the monitoring

programme were widened in 2001, eval-

uations in the fi eld of biodiversity have

been intensifi ed. Chapter 4 presents

interim results on the relationships

between ground vegetation and envi-

ronmental factors. Once again, evalu-

ations are continuing and more details

are expected in the coming years.

Co-operating partners were invited

to illuminate the context of the on-

going work. Th e contributions of the

ICP on Modelling and Mapping and of

the European Environmental Agency

are included in Chapters 3 and 4, re-

spectively.

Intensive Monitoring Plot, Ireland.

Surveys conducted Level I Level II

Crown condition annually annually all plots

Foliar condition once so far1 every 2 years all plots

Soil chemistry once so far2 every 10 years all plots

Soil solution chemistry - continuously some plots

Tree growth - every 5 years all plots

Ground vegetation - every 5 years some plots

Atmospheric deposition - continuously some plots

Ambient air quality continuously some plots

Meteorological condition - continuously some plots

Phenology (optional) - according to phenophases some plots

Remote sensing (optional) - some plots 1 on 1497 plots | 2 on 5289 plots Table 1-1: Surveys carried out on Level I and Level II plots.

Figure 1.1: Flow diagram illustrating the relationships between site and stress

factors and the forest ecosystem condition. Boxes and arrows in bold are spe-

cifi cally investigated in this year’s report.

. C

2.1 Introduction

Th e annual crown condition survey

is the main tool of the programme to

obtain a large scale overview on the

condition of forests in Europe. In 2001

the assessments were conducted in all

EU member states and in 15 non-EU

countries on the transnational 16 x 16

km grid. Some 132 000 trees were as-

sessed on nearly 6 000 plots during the

summer months. Quality assurance

measures were routinely applied in

the countries and extensive plausibil-

ity and consistency checks were car-

ried out by the Programme Co-ordi-

nating Centre in Hamburg, Germany.

Due to changes in the assessment meth-

ods French and Italian datasets were

excluded from the time series.

Th e crown condition is assessed in

terms of defoliation. Th is parameter

describes the lack of foliage for each

sample tree. Defoliation depends on

many stress factors and is therefore a

valuable measure to describe the over-

all forest condition.

2.2 Crown condition in 2001 and

trends

22.4% of all trees assessed in 2001 were

classifi ed as moderately or severely de-

foliated or dead. Crown condition in

the EU Member States was slightly

better than in Europe as a whole. Of

the four tree species most frequently

occurring on the plots, European and

sessile oak were still the most severe-

ly defoliated species and also showed

the highest proportion of dead trees

(Fig. 2-1).

Th e temporal development of defoli-

ation was analysed for a sample of all

continuously monitored trees. With

the exception of the holm oak, mean

defoliation of all main tree species in-

creased in 2001 (Fig. 2-2). Th e share of

damaged and dead trees (defoliation

classes 2-4) of all species was high-

est in 1995 (25.8%) and decreased in

the following two years. Since then a

steady but slow increase in damage has

been recorded.

Th e mid-term development of defolia-

tion not only varies between tree spe-

cies, but also between regions. Th e plot-

wise mapping (Fig. 2-3) shows that the

number of plots with a signifi cant in-

crease (565) is slightly larger than the

number of plots with an decrease of

mean defoliation (500).

Regions with prevailing improvements

of crown condition are southern Poland

and south-western France. Deteriora-

tion took place mainly in eastern Bul-

garia and southern Italy. Local experts

explain the observed deterioration in

southern Europe mainly by unfavour-

able weather conditions. For Bulgaria,

extensive forest fi res were also report-

ed and the damaged areas in southern

Italy are among the regions with the

highest ozone concentrations in Europe

during the observation period. In addi-

tion, beech and chestnut plots suff ered

from severe insect and fungal attacks.

Th e improvement in southern Poland

is ascribed to a reduction of air pollu-

tion emissions and favourable weath-

er conditions, especially in the period

from 1994 to 1999.

Corsican pine, France.

S • More than 20% of 132 000 trees

assessed in 2001 were classifi ed as

damaged.

• Trees that have been monitored

since the start of the survey show

continuous deterioration from 1986

to 1995. After a marked recupera-

tion in the mid-1990s the deteriora-

tion resumed at a lower level.

• In-depth evaluations for Scots pine

and common beech show that there

is no uniform trend of defoliation

throughout Europe. Rather, they

reveal changing conditions in dif-

ferent regions.

• High or low precipitation, insect

and fungi attacks as well as air pol-

lution are among the most impor-

tant infl uencing factors. Whereas

sulphur deposition correlates with

high defoliation, nitrogen has am-

biguous eff ects: depending on soil

and forest type, either its fertilis-

ing or its acidifying eff ects seem to

prevail.

Tree crowns of undamaged, moderately damaged and dead Scots pine.

Figure 2-2: Development of mean de-

foliation for European main tree

species, calculated for continuous-

ly monitored trees. Sample sizes vary

between 1 215 trees for European

and sessile oak and 3 012 for spruce.

Figure 2-1: Percentage of trees in diff erent defoliation classes for main tree species (-groups).

Total Europe and EU, 2001.

EU Member States Total Europe

% o

f tr

ees defoliation

class

needle/leaf

loss

degree of

defoliation

up to 10%

all s

pec

ies

conif

ers

broa

dle

aves

all s

pec

ies

conif

ers

broa

dle

aves

Nor

way

spru

ceSc

ots

pin

eC

omm

. bee

chEuro

p.+

Ses

s. o

ak

dead

severe

moderate

none

Scots pine Norway spruce beech

European+sessile oak holm oak maritime pine

mea

n d

efo

liat

ion

%

slight

2.3 Infl uences on crown condition

Figure 2-3: Development of defolia-

tion for all species.

Plot wise linear trends for 1994 –

2001 were tested for signifi cance. Th e

evaluation period for France, Italy

and Sweden is 1997 to 2001.

Th e data and the statistical analysis

• Data basis: In-depth evaluations for

Scots pine and beech are based on

those Level I plots for which data

on at least three pine or beech

trees were continuously reported

from 1994 to 1999. Th e evaluation

period ended in 1999 as later dep-

osition data was not available.

• Levels of defoliation: Defoliation

fi eld estimates throughout Europe

are strongly infl uenced by stand age

(older trees are usually more defo-

liated) and by the country in which

the Level I plot is located (assess-

ment methods sometimes vary be-

tween countries). Th e levels of de-

foliation presented were therefore

evaluated as diff erences between

fi eld estimates and modelled plot

values which take into account the

variables ‘stand age’ and ‘country’

and hence compensate for their in-

fl uence.

• Th e development of defoliation

was calculated as the plot wise lin-

ear gradient of a regression through

all annual mean plot values of the

years 1994 to 1999. Age and country

infl uences were negligible for time

trend evaluations.

• Th e geostatistical method kriging

was used to interpolate levels and

trends of defoliation, based on the

available Level I plots. Interpolat-

ed values were only calculated for

grid points with more than 4 plots

available in a radius of 100 km.

• Multivariate models were used to

explain defoliation by diff erent en-

vironmental infl uences. A coinci-

dence of high defoliation with cer-

tain stress factors can be interpreted

as damaging eff ect.

Figure 2-4: Diff erences between me-

dium term mean defoliation of Scots

pine and model value; interpolation

based on 1313 plots continuously as-

sessed from 1994 to 1999.

Figure 2-5: Linear time trends of

mean defoliation of Scots pine; inter-

polation based on 1313 plots contin-

uously assessed from 1994 to 1999.

Figure 2-6: Diff erences between me-

dium term mean defoliation of

common beech and model value;

interpolation based on 399 plots con-

tinuously assessed from 1994 to 1999.

Figure 2-7: Linear time trends of

mean defoliation of common beech;

interpolation based on 399 plots con-

tinuously assessed from 1994 to 1999.

Scots pine

In Estonia, southern Poland as well as

north eastern Spain there are regions

with a comparatively high mean de-

foliation. However, crown condition

has improved in these regions (Fig.

2-4; 2-5). Also, in middle Norway a

decrease of the rather high mean de-

foliation has been observed, whereas

in southern Norway the comparative-

ly good crown condition has deterio-

rated. Th e deteriorating trend in Bul-

garia is based on the limited number

of 21 pine plots of which 19 show a

worsening trend.

Common beech

Southern Germany shows a compara-

tively high mean defoliation of beech

which has worsened towards the end

of the observation period (Fig. 2-6; 2-

7). Romania is characterised by obvi-

ously high fl uctuations in beech crown

condition. Th e high defoliation in cen-

tral Romania has decreased until 1999

whereas the comparatively low defolia-

tion in the middle east and middle west

of the country has clearly increased.

Other European regions with dete-

riorating crown condition for beech

trees are north-western Germany and

the region along the border between

Slovenia and Croatia. Improvements

have been registered for Slovakia and

regions in Germany.

signifi cant decrease

no change

signifi cant increase

percentage of plots

Defoliation of Scots pine

signifi cantly below model

below model

model

above model

signifi cantly above model

Development of Scots pine

signifi cant improvement

improvement

no change

worsening

signifi cant worsening

Defoliation of common beech

signifi cantly below model

below model

model

above model

signifi cantly above model

Development of common beech

signifi cant improvement

improvement

no change

worsening

signifi cant worsening

Multiple infl uences on crown con-

dition

Model results show that high precip-

itation is related to relatively healthy

tree crowns (Tab. 2-1). In addition, pine

plots show a plausible interaction of

site characteristics and precipitation:

on plots with low and medium water

availability there is a positive correla-

tion between precipitation and crown

condition. It seems that on these plots

an increased water supply improves

forest condition, whereas the reverse

is true for sites with more than enough

water available in the soil. With respect

to biotic damage factors, insects (and

on beech plots also fungi) are related

to high or increasing defoliation. Sul-

phur deposition was also correlated in

all four models with high or increas-

ing defoliation. Research results on the

damaging eff ects of sulphur inputs are

thus supported. Th e correlations be-

tween nitrogen inputs and forest condi-

tion are not signifi cant and reveal am-

biguous conditions. Th is might confi rm

current knowledge, as nitrogen inputs

on one hand eutrophy forest ecosystems

but on the other hand may have acidi-

fying eff ects. Interaction terms of dep-

osition and soil pH in the model (not

depicted) show that eff ects of deposi-

tion in general depend on the acidity

status of the soil. A linear trend could

explain parts of the temporal varia-

tion of defoliation for pine as well as

for beech. Th is shows that there are

linear trends which are independent

from the other explanatory variables

of the model. As already shown on the

maps, however, there is not a uniform

Europe-wide trend, but varying con-

ditions on diff erent plots.

F :

UNECE EC. . L, M., M,

V., B, G., S, W., F, R.,

L, D., D, D., B,

U.: T C F E-

. T R. UNECE

EC, G B, .

DefoliationR2

No.

of

plots

Variables

precip

index

site*precipa

interactioninsect fungi

depositionblinear

trendage country

S NHx

NOy

Spatial

variation

pine 60.9 1313 - * + + + - * *

beech 41.1 399 - + + + - + * *

Temporal

variation

pine 44.5 1313 - * + + + - *

beech 39.3 399 - + + + - + *

-negative

correlation -signifi cant negative

correlation +positive

correlation +signifi cant positive

correlation * correlation *signifi cant

correlation

a data source: Global Precipitation Climatology Centre (www.dwd.de/research/gpcc)

b data source: EMEP 150 x 150 km grid (www.emep.int)

Table 2-1: Relations between temporal and spatial variation of defoliation of Scots pine and common beech and var-

ious explaining variables as results of multivariate regression analyses. Th e R² value indicates the percentage of vari-

ance explained by the model.

Introduction

Storm damage is one of the most im-

portant economic factors of forest dam-

age in Europe. Over the past decades,

damage severity has increased. Th e

storms in December 1999 caused the

highest damage ever reported in Europe

(nearly 200 million m³ merchantable

timber). Th e severity of the storms is

refl ected in the damage that occurred

on the plots of ICP Forests and EU: 3%

of all Level I plots and 12% of the Lev-

el II plots were damaged. In the main

storm areas these percentages were

much higher. In France and Switzer-

land the monitoring plots were used

to investigate the factors that contrib-

uted to the damage in 1999. In general

the results are in line with earlier eval-

uations carried out in Germany, Aus-

tria and other countries after the se-

vere storms in 1990.

Wind exposure

In France and Switzerland wind speed

was linked to increased damage. For-

ests growing in the plains or on gentle

slope suff ered the highest damage.

Tree species

For both countries earlier fi ndings were

confi rmed showing that conifers are in

general more susceptible to wind than

deciduous trees. During winter ever-

green conifers off er more resistance to

wind than deciduous trees, which in-

creases their vulnerability to storms.

Stand structure and management

Storm damage increased substantially

with tree height. For the French plots,

higher ratios of stand height to stand

mean diameter meant higher dam-

age for some species. Unevenly struc-

tured stands seemed to be more resist-

ant than single or multi-storey stands

in Switzerland. Th inning increased the

susceptibility to wind damage in the

fi rst few years.

Previous damage and biological

condition

In Switzerland, stands damaged by the

1990 storms were also more often dam-

aged by the 1999 storms, indicating ei-

ther a higher susceptibility of these sites

or increased vulnerability to wind as a

result of the prior storms. Crown de-

foliation was not found to be a signif-

icant factor. Root and stem rot seem

to increase the chances of uprooting

or stock-breakage of Norway spruce in

Switzerland and red oak in France.

Soil condition

Water saturated and water-logged soils

increased the danger of storm damage

in Switzerland, whereas the stability of

stands increased with rooting depth in

France. Th e infl uence of soil acidifi ca-

tion and increased nitrogen deposition

is currently being investigated.

Consequences and outlook

Since the 1999 storms, bark beetle pop-

ulations have increased and the mortal-

ity of susceptible tree species, such as

Norway spruce and various pine spe-

cies, is expected to rise in many of the

aff ected areas in the coming years.

Total area and stocking volume of Eu-

ropean forests and the proportion of

older and taller stands are current-

ly increasing due to change in man-

agement practice. Although ecologi-

cally benefi cial, this will increase the

vulnerability of forests and therefore

also the risk of wind damage in the fu-

ture. Th e role of climate change and

possibly increasing storm intensities

and frequencies is still unclear. How-

ever, various scenarios indicate a high-

er probability of unfavourable weath-

er conditions such as extreme wind

speeds and heavy rainfalls.

T

Storm damage in France.

after 1994 with more than 20% of the

trees damaged in some years (Fig. 2-8).

Recuperation has been observed since

1997, although the level of damage is

higher than at the beginning of the ob-

servation period.

Th e decline process

Th e oak decline can be observed at var-

ying degrees of intensity and, mostly

without one single identifi able cause.

For holm oak stands, three types of de-

cline processes have been identifi ed:

• Sudden death, when an apparent-

ly healthy tree without any signs of

damage dies in a short period of

time;

• Progressive decline, when the stand

shows debility symptoms, losing

crown density, and at the same time,

showing dead branches and twigs.

Th is process mostly ends with the

death of trees with a decline period

lasting from two to several years;

• Loss of vitality of trees, the symp-

toms are the same as in the case of

progressive decline, nevertheless the

trees continue to survive, though in

a decrepit state.

Causes of damage are assessed in the

European wide Level I survey and addi-

tional national surveys like the Spanish

Forest Damage Inventory (IDF).

Results show a complex of predispos-

ing, triggering and ancillary causes of

damage. Th e damage is often infl uenced

by soil type and soil capacity to retain

water. Climatic factors, in particular

drought stress caused by low precipita-

tion and/or warm temperatures, often

trigger outbreaks of oak decline. Inap-

propriate human intervention ranging

from complete absence of silvicultur-

al management to abusive use of the

resources may have increasing eff ects.

Once the process has started, a lot of

opportunistic biotic agents can speed

it up, sometimes leading to the death

of a tree or tree group within a short

period of time.

Th e most common biotic agents in-

clude fungi (Phythophtora, Armillar-

ia, Ophiostoma, Cryphonectria, Fusar-

ium, Biscogniauxia), insects (Platypus,

Coroebus, Cerambix, Lymantria, Tor-

trix), bacteria (Brennia), nematodes and

viruses.

Th e interplay of harmful factors as de-

scribed above is a simplifi ed scheme.

In many cases a diff erent course of the

decline can be observed and there are

cases when the deteriorating crown

condition or death of holm oak trees

remains unexplained.

Th e only possible action in the mid

term seems to be the implementa-

tion of tried and tested silvicultural

and conservative management tech-

niques including an even closer sur-

vey of the holm oak stands in order to

enable timely interventions and sani-

tary fellings. Also, it will become in-

creasingly important to create stable

stands, including mixed stands. Th is

will hopefully minimise the impact of

the oak decline and strengthen the spe-

cies’ resistance to adverse environmen-

tal conditions.

T

(Q )

Holm oak stands are among the most

characteristic elements of the Mediter-

ranean vegetation. Together with other

oak forests they are very complex and

diverse natural ecosystems in Europe.

Th eir natural range spreads from Por-

tugal to Turkey. From northern Spain

and southern France they reach south to

the African Magreb. Th ey can be found

from sea level to altitudes of more than

1 500 m (2 800 m in Morocco).

Holm oak is regarded as undemand-

ing as far as soil is concerned. It nor-

mally shows an extraordinary resist-

ance to drought, continental conditions

and very dry winds. It is known as a

slow growing species and an eff ective

colonizer of marginal soils where pure

and dense stands can be found, which

are very typical for the Mediterrane-

an landscape. Th e species reproduc-

es very successfully from seed, and its

root and stump sprouting capability

remains active even after the age of

200 years. Holm oaks can live to more

than 700 years.

Th e species and the forests it forms are

closely connected to the cultural devel-

opment of the Mediterranean region.

Its longevity, adaptability and the va-

riety of products that it is able to pro-

duce, together with the great capability

to protect soil and ground vegetation,

were the basis for what can be consid-

ered as the fi rst historical example of

sustainable forest management: the

“dehesa” in Spain and the “monta-

do” in Portugal. Th ese are ecosystems

with a low density of trees (but good

crown coverage of soil) under which a

fairly sparse maquis and herbal natu-

ral species develop together, creating

ideal conditions for rich animal wild-

life. Th eir essential functions in terms

of nature and landscape preservation

recently led to the development of of-

fi cial codes for the defence and pro-

motion of these ecosystem types. In

other countries holm oak occurs in

dense stands forming semi natural

forests managed mostly for protec-

tive functions.

In the last 10 years, 7 countries have

monitored holm oak on Level I plots

(Romania, Greece, Croatia, Italy,

France, Spain and Portugal). Most plots

are located in Spain, with almost 3 200

trees, and Portugal, with more than 650

trees. Th e development of the percent-

age of damaged trees (defoliation class-

es 2 + 3) shows a signifi cant increase

S • Holm oak forests are important

elements of the Mediterranean

vegetation.

• Th ey are characterised by a wide

ecological range and a high adapt-

ability, by their biodiversity rich-

ness and protective value. Th e great

variety of products which they off er

and the historical links to the Medi-

terranean culture make holm oak

forests an essential forest formation

of this region.

• A process of decline detected

throughout its natural range seems

to be caused by the combined eff ects

of climatic, biotic and human fac-

tors.

„Dehesa“ - „Montado“ forest forma-

tions.

Closed holm oak forest, Spain .

Percentage of damaged holm oak (de-

foliation classes 2+3) in Europe and

Spain.

Sudden death of holm oak.

Progressive decline of holm oak.

Europe

Spain

% d

amag

ed t

rees

. P

,

3.1 Introduction

For most European countries critical

load maps for nitrogen and acidity are

available based on estimated data (see

box on p. 20). Th e large number of Lev-

el II plots, their comparatively wide ex-

tend and the extensive database off er

the possibility to validate and improve

existing models and to contribute to

the development of new methods. A

Europe-wide assessment of critical

loads based on measured data of In-

tensive Monitoring Plots in compari-

son with measured present loads has

not yet been available and is presented

for the fi rst time in this report. As data

collection, submission and validation is

rather time consuming, data up to 1999

were used. Evaluations were conduct-

ed after intensive checks on data reli-

ability and consistency. Critical loads

were calculated for approximately 230

Intensive Monitoring Plots where all

relevant data on deposition, meteor-

ology, forest growth and soil and soil

solution chemistry were available. Re-

sults for nitrogen are reported as the

sum of nitrate (NO ₃ ) and ammonium

(NH ₄ ). Acidity is defi ned as the sum of

sulphate (SO ₄ ) and nitrogen.

Defi nition of critical limits and

loads

Atmospheric inputs aff ect diff erent

parts of the forest ecosystems simul-

taneously. Th erefore various related

critical loads can be calculated that

take into account these diff erent ef-

fects. Th e lowest of these is the critical

load relevant for the protection of the

specifi c ecosystem. It has to be taken

into account that the presented critical

loads refer to a steady state. An excess

implies an increase in the concentra-

tion of nitrogen and acidity ultimately

reaching the critical limit. In practice

the time before eff ects become visible

can take several years to decades.

In this report critical loads for nitrogen

were calculated which aim at no fur-

ther net accumulation of nitrogen in

the soil. Th e calculations are based on

a nitrogen threshold in the soil solution

of 0.28 g.m ³ (0.02 mol c .m ³ ). For sites

with higher values, increased leaching

is to be expected. In addition, critical

nitrogen loads are included which aim

at ensuring that concentrations of ni-

trogen in the foliage of trees stay below

a critical limit of 18 g.kg ¹ . Above this

limit, eff ects on trees can be expected

such as an increased vulnerability to

drought stress, frost, pest and diseas-

es. Another approach aims at deter-

mining eff ects of nitrogen deposition

on ground vegetation. Here, the limits

are based on empirical data.

Th e critical loads for acidity take into

account the impact on tree roots of

free aluminium in the soil solution.

Th ey were calculated by aiming that

ratios of toxic aluminium to base cat-

ions in the soil solution stayed below a

critical limit of 0.8 for pine and spruce

and 1.6 for oak and beech. Other criti-

cal loads for acidity assume no further

loss of exchangeable base cations in

base rich forest soils (loess, clay and

peat soils) and no further loss of read-

ily available aluminium in base poor

sandy forest soils.

Critical loads for heavy metals were cal-

culated which ultimately lead to con-

centrations in soil solution that may

aff ect soil organisms. For cadmium a

concentration of 0.8 mg.m ³ was used;

for lead the limit was 8 mg.m ³ .

3.2 Nitrogen

Th e average nitrogen deposition

from 1995 to 1999 on all 234 plots is

19 kg.ha ¹ .yr ¹ . Lowest loads were

found for pine, followed by spruce,

refl ecting their location in mostly low

deposition areas, such as Scandinavia

(Tab. 3-2). High nitrogen inputs above

22.4 kg.ha ¹ .yr ¹ (1 600 mol c .ha ¹ .yr ¹ )

only occur on plots in central Europe

(Fig. 3-1). Total nitrogen input is gen-

erally found to be much lower on plots

in northern and southern Europe.

Soil profi le, formulas.

S • Th e Level II plots off er an unri-

valled chance to validate and

improve existing models for the

calculation of critical loads. Such

loads have been calculated refer-

ring to diff erent parts of the forest

ecosystem.

• Critical loads for nitrogen aiming

at no further nitrogen accumula-

tion in the soil are exceeded at

92% of the investigated plots.

Critical loads taking into account

eff ects on trees are exceeded at

45% of the plots. At these plots an

increased vulnerability to drought

stress, frost, pest and diseases is to

be feared. Th e plant diversity of the

ground vegetation is potentially en-

dangered at 58% of the plots.

• For 33% of the investigated plots

critical loads for acidity referring

to eff ects on tree roots are exceeded.

Leaching of base cations or alumin-

ium already occurs at lower acidity

inputs. Related critical loads are

exceeded at 64% of the plots.

• In general, highest excess acid depo-

sitions occur in central Europe and

for nitrogen also in western Europe,

where present loads are high and

critical loads are relatively low.

• Critical loads of cadmium and lead

were exceeded on 91% of the plots

for lead and on 29% of the plots for

cadmium. Th ese results are based

on very stringent criteria for pos-

sible impacts on soil organisms.

Critical loads and levels – a tool

for environmental policy

Critical loads and levels defi ne thresh-

olds for the eff ects of air pollution. If

pollution is below the critical values, it

is assumed that no environmental dam-

age will occur and a long-term stability

of the ecosystem is achieved. Th e crit-

ical load of sulphur and nitrogen acid-

ity was defi ned in 1994 in the UNECE

Protocol on Further Reduction of Sul-

phur Emissions:

“Critical Load” means a quantita-

tive estimate of an exposure to one or

more pollutants below which signifi cant

harmful eff ects on specifi ed sensitive el-

ements of the environment do not occur,

according to present knowledge.

Direct eff ects on plants, due to mostly

gaseous concentrations of air pollut-

ants, are defi ned in a similar way us-

ing critical levels.

Critical loads are derived by compar-

ing the quantity of mainly anthropo-

genic pollutants as inputs on one side

and the removal, acceptable storage and

outputs of these substances on the oth-

er side. Th e outputs include the harm-

less or tolerable transfer of the pollut-

ants to other parts of the environment.

Critical loads are not exceeded as long

as the system remains in balance, but

any additional input of pollutants may

cause harmful eff ects. Th e comparison

between critical loads and the actual

deposition makes it possible to deter-

mine excess deposition values. Map-

ping the extent of excess depositions

for given receptors provides an im-

portant policy tool for the develop-

ment of optimised pollution abate-

ment strategies.

An eff ective policy tool

Two eff ects-based protocols, the 1994

Sulphur Protocol and the 1999 Gothen-

burg Protocol were adopted under the

UNECE Convention. Th e 1999 Gothen-

burg Protocol aims at curbing the emis-

sions of sulphur, nitrogen and volatile

organic compounds and is an example

for the use of critical load maps. For the

most recent maps, 24 countries calcu-

lated and submitted data estimates for

1.3 million gridcells to the Internation-

al Co-operative Programme (ICP) on

Modelling and Mapping, which were

used to produce the latest maps on crit-

ical loads for forests, crops, natural veg-

etation, soils, water and materials.

Th e current critical loads methodology

describes a steady-state condition, and

thus aims towards a long-term stabili-

ty of ecosystems. In order to map and

evaluate present damage as well as pre-

dicted changes it is important to ap-

ply dynamic modelling. Recently, ICP

Modelling and Mapping developed a

dynamic modelling approach in collab-

oration with ICP Forests to be applied

on the Level II plots. It is anticipated

to complete the eff ort on 200 forest

plots by 2004 and to up-scale it to the

whole of Europe afterwards.

F :

://..

Sensitivity to sulphur expressed in

terms of critical loads. Red indicates

areas with ecosystems most sensitive

to acid deposition, whereas areas in

blue are least sensitive.

Excess deposition of acidity over crit-

ical loads (CL) in 1990 (left) and

2010 (right) if the Gothenburg proto-

col is fully implemented. Th e scenar-

io shows a decrease in excess depo-

sitions due to a substantial sulphur

reduction. However, the emission of

nitrogen oxides and ammonia will

continue to contribute to acidifi ca-

tion and eutrophication.

Th e average critical load aiming at no

further nitrogen accumulation in the

soil was near 8 kg.ha ¹ .yr ¹ (Fig. 3-1).

Th ese critical loads were exceeded on

92% of the evaluated Level II plots (Tab.

3-1 and 3-2). Critical loads are lower

for pine with a lower nitrogen uptake,

than for the other tree species. High

critical loads characterise ecosystems

which are less sensitive to high nitro-

gen inputs. Such plots are mainly lo-

cated in southern Europe, where for-

est ecosystems, specifi cally broadleaf

forests, have a higher nitrogen uptake.

Results confi rm that forests in northern

Europe are more sensitive to nitrogen

inputs as the net uptake of nitrogen by

trees is low in these regions.

Critical nitrogen loads related to eff ects

on tree foliage were higher. Th us reac-

tions of trees are expected at higher

nitrogen inputs only. Th e average was

near 14 kg.ha ¹ .yr ¹ for pine and near

20 kg.ha ¹ .yr ¹ for spruce. Th ese loads

were exceeded at 45% of the evaluated

conifer plots indicating an increased

vulnerability to drought stress, frost,

pests and disease.

Critical loads requiring no changes in

the ground vegetation were exceeded

on 58 % of the plots. Th is shows that

changes in plant diversity are very like-

ly in European forests.

Deposition measurements and

units

Within the Intensive Monitoring

Programme atmospheric deposition

is measured below the forest cano-

py (throughfall), directly at the tree

trunks (stemfl ow) and in nearby open

fi elds (bulk deposition). Total deposi-

tion on forest stands was derived by

adding throughfall and stemfl ow val-

ues, while correcting for the eff ects of

element interactions with the canopy

(uptake and leaching through leaves

and needles). Th is correction requires

data on bulk deposition.

Acidity is given in mol c .ha ¹ .yr ¹ (read

“mol of charge per hectare per year”).

Chemicals are usually deposited into

soils in the form of charged particles

or ions. Ions can carry diff erent charg-

es and the chemical reactions in the

soil depend among other things on

the charges received. Mol of charge

is a way of expressing the total charg-

es deposited into the ecosystem and

allows comparisons of the deposition

of diff erent substances. Th e simpler

unit of kg.ha ¹ .yr ¹ does not allow this

comparison. For example: 1000 mol c

is equivalent to 14 kg of nitrogen and

16 kg of sulphur.

Th roughfall sampler in a beech stand,

Belgium.

Ecosystem compart-

ment concerned

% of Level II plots with excess of

critical nitrogen

loads

critical acidity

loads

Soil 92 64

Tree 45 33

Ground vegetation 58 -

Table 3-1: Percentage of Level II plots

with deposition above critical loads

related to diff erent compartments of

the forest ecosystem.

Figure 3-1:

Top map: average present deposition load of nitrogen (N = NH ₄ -N plus NO ₃ -N)

Middle map: critical nitrogen loads related to nitrogen concentration in the soil.

Bottom map: excess deposition above critical loads.

234 Intensive Monitoring Plots, average 1995-1999.

Table 3-2: Average total present nitrogen deposition load

(PDL kg) and critical load (CL kg) in kg.ha ¹ .yr ¹ as well as

percentage of plots with excess deposition (CLex %). Criti-

cal loads are related to eff ects in the soil. Values in brackets

are given in mol c .ha ¹ .yr ¹ .

SpeciesNo. of

sitesPDL kg CL kg CLex (%)

Pine 57 15 (1074) 6 (419) 96

Spruce 96 19 (1359) 9 (618) 86

Oak 28 21 (1476) 9 (623) 93

Beech 42 22 (1540) 9 (659) 98

Other 11 17 (1198) 9 (670) 91

All 234 19 (1329) 8 (580) 92

Present Deposition Load

Critical Load

Critical Load Excess

3.3 Acidity

Th e average acid load (nitrogen plus

sulphate) on 226 plots is nearly 2 100

mol c .ha ¹ .yr ¹ . As with nitrogen, lowest

loads were found for pine, followed by

spruce (Tab. 3-3). Relatively high acid

inputs can be found everywhere in Eu-

rope, except in central and northern

parts of Scandinavia, but most sites

with highest acid inputs (up to 3 000

mol c .ha ¹ .yr ¹ ) are situated in central

Europe (Fig. 3-2).

Critical loads, which take into account

the impact on tree roots through free

aluminium in the soil solution, are

clearly lower for pine and spruce. Th ese

species are more sensitive to alumini-

um than oak or beech. In general, the

critical acid load increases from the

northern boreal regions to southern

Europe, which shows that forest eco-

systems in the south are less sensitive to

acidic inputs. Th is is fi rstly due to high-

er neutralising base cation inputs from

the atmosphere and from soil weather-

ing and secondly to a higher nitrogen

uptake by the vegetation in the south.

Th e critical loads are exceeded at 33%

of the plots (Tab. 3-1 and 3-3).

Critical loads related to base cation

and aluminium pools in the soil are

lower. Th ey are exceeded at 66% of

the plots.

3.4 Heavy metals

On average the present lead deposition

is much higher than the critical load,

whereas the excess is small for cad-

mium. Th e share of plots where crit-

ical loads were exceeded was 91% for

lead and 29% for cadmium (Tab. 3-4).

Th ese results are, however, based on

very stringent criteria related to pos-

sible impacts on soil organisms.

3.5 Outlook

Apart from critical loads, present depo-

sition thresholds were calculated. Th ey

take into account the present plot spe-

cifi c situation and do - in contrast to

critical loads - not assume a steady-

state. Th us they are an even more pre-

cise instrument to evaluate risks to the

forests. Preliminary calculations indi-

cate that present deposition thresholds

are higher than critical loads for ni-

trogen, whereas the reverse is true

for acidity. Th is aspect will be consid-

ered in greater detail in next year’s Ex-

ecutive Report, when dynamic mod-

els will be applied to predict impacts

of acid deposition scenarios on forest

soils. Th e further development of crit-

ical loads needs a continuation of the

close co-operation with other bodies

and programmes under the Conven-

tion on Long-range Transboundary

Air Pollution.

F :

UNECE EC, ; D V, W., G.J.

R, H. D, D. Z, D.

A, P. N, M. P, J. A,

J.C.H. V E. V. I

M F E

E. T R . UNECE

EC, G B, .

Figure 3-2:

Top map: average present deposition load of acidity (SO ₄ + NH ₄ + NO ₃ )

Middle map: critical loads of acidity related to eff ects on tree roots

Bottom map: excess depositions

226 Intensive Monitoring Plots, average 1995-1999.

Table 3-3: Average total present acid deposition load (PDL

mol) and critical load (CL mol) in mol c .ha ¹ .yr ¹ as well as

excess deposition (CLex %). Critical loads are related to ef-

fects on tree roots.

SpeciesNo. of

sitesPDL mol CL mol CLex (%)

Pine 55 1749 2906 40

Spruce 94 2146 2726 34

Oak 27 2272 4721 25

Beech 40 2346 4624 31

Other 10 2032 5282 18

All 226 2094 3469 33

Number of

sites for all

tree species

PDL (g.ha¹.yr¹) CL (g.ha¹.yr¹) CLex (%)

lead cadmium lead cadmium lead cadmium

242 26 0.38 3.8 0.33 91 29

Table 3-4: Average total present dep-

osition load (PDL), critical load (CL)

and excess deposition (CLex excess)

of lead and cadmium (in g.ha ¹ .yr ¹ ).

Present Deposition Load

Critical Load

Critical Load Excess

. G

4.1 Introduction

Th e species composition of the ground

vegetation assessed at Intensive Moni-

toring Plots is an indication of the plant

diversity of forest ecosystems. Level II

off ers a unique opportunity to relate the

species composition of the ground veg-

etation to environmental factors, in-

cluding atmospheric deposition. Th is

was done to identify where possible

those environmental factors that most

strongly determine the plant diversi-

ty of the ground vegetation, specifi -

cally in view of the hypothesis that the

deposition of nitrogenous compounds

is an important threat. If such factors

are known, it may be possible to assess

more precisely threats to plant diversi-

ty, to which local governments might

respond proactively.

4.2 Plant diversity at plot level

An evaluation of plant diversity, indi-

cated by the Simpson index, was car-

ried out with the available data from

674 plots (Fig. 4-1). Th e value of this

index is higher when more species

occur. Th e results show that there are

large diff erences on the plots through-

out Europe.

Cranberries and lichens, Finland .

S • Ground vegetation records from

674 plots off er a unique chance for

biodiversity evaluations and reveal

large variations in plant diversity

throughout Europe’s forests.

• Plant diversity was found to be

less in acid conditions which are

now widespread in Europe. Actual

soil acidity is clearly infl uenced by

previous depositions.

• Nitrogen deposition had a small

but signifi cant direct infl uence on

ground vegetation.

• Nutrient rich soils, southern cli-

mates, and oak stands are corre-

lated with diverse and species rich

ground vegetation.

• Future data collection will allow

a more appropriate assessment of

deposition impacts on vegetation

changes.

Biodiversity in European forests

Forests, the natural potential land-cov-

er in most of Europe, off er diverse hab-

itats for plants, animals and micro-or-

ganisms and hold the vast majority of

the terrestrial species. Many species

contribute towards the functioning of

the forest ecosystem, including such di-

verse groups of organisms as mycor-

rhizal fungi, herbs, trees and insects.

Forests with a good biodiversity status

are considered to particularly provide

ecological, economic and social func-

tions and benefi ts to the society. Th ey

maintain the water balance, and pro-

tect soils against erosion. Today, in re-

gions of Europe with long industrial

and agricultural heritage and high pop-

ulation density forest cover has most-

ly been reduced. Furthermore, also in

the remaining forest regions, forest-

ry and other uses of forests may have

impoverished the biodiversity. Espe-

cially the disappearance of old growth

forest has signifi cantly contributed to

this loss. On the other hand, traditional

forest use has not generally been neg-

ative from a biodiversity point of view

and even created specifi c ecosystems

that nowadays are considered to have

particular biodiversity values. Exam-

ples are forms of coppicing and forest

grazing systems, uneven-aged mixed

forests in central Europe, and Mediter-

ranean agro-forestry systems.

F :

://..

Dead wood not only provides food

and shelter for numerous insect and

fungi species, it can also be impor-

tant for the growth of new young

trees.

Figure 4-1

Diversity of vascular plants. Case

studies at 674 unfenced Intensive

Monitoring Plots, assessed in 1998-

1999.

Structured forests provide a great

range of habitat types. Th e depict-

ed intensively managed mixed moun-

tain forest off ers high ecological, eco-

nomic and recreational values.

International processes and

projects

Biodiversity was a priority issue at the

UN Conference on Environment and

Development in Rio de Janeiro 1992

resulting in the “Convention on Bio-

logical Diversity” (CBD) and a set of

“Forest Principles”. In this convention

biodiversity is defi ned as

“the variability among living organisms

from all sources including inter alia,

terrestrial, marine and other aquat-

ic ecosystems, and the ecological com-

plexes of which they are part; this in-

cludes diversity within species, between

species and of ecosystems.”

In a follow up, most European coun-

tries, and also the EU, have elaborated

biodiversity strategies and plans and

have integrated the biodiversity con-

siderations in their legal framework (in

EU e.g. in the “Habitats Directive” an

the “NATURA 2000” ). Activities in line

with the Forest Principles have most

notably been implemented in regional

processes of which one is the “Minis-

terial Conference of Protection of For-

ests in Europe” (MCPFE) .

A necessary fi rst step in the assess-

ment of biodiversity is to develop val-

id and cost-eff ective indicators. Related

projects funded by the European Un-

ion include “Indicators for monitor-

ing and evaluation of forest biodiver-

sity in Europe” (BEAR) , the “European

Biodiversity Assessment Tools” (Bio-

Assess) and the “Nature-based manage-

ment of beech in Europe” (Nat-Man) .

Additional research projects address

various aspects of forest genetic bio-

diversity.

Ideally, eff orts should be coordinat-

ed. Without doubt, the pan-Europe-

an monitoring system of ICP Forests

and the EU collects data that have the

potential to contribute to the assess-

ment of air pollution eff ects on forest

biodiversity in Europe.

F :

://-...

Figure 4-2: Response curves show-

ing the probability of occurrence of

36 species at diff erent pH. Th e spe-

cies can be grouped into three species

classes with an optimum at low, in-

termediate, and high pH; assessed in

the period from 1998 to 1999 for 366

Intensive Monitoring Plots.

Deschampsia fl exuosa forms typical

grass vegetation in open pine stands

on acid soils.

Galium odoratum is a characteristic

species in many beech and mixed de-

ciduous forests.

Ajuga reptans growing on a moist

and base cation rich site.

Relationships between plant diversi-

ty and species numbers of the ground

vegetation on one hand and environ-

mental factors on the other hand were

evaluated for approximately 200 plots

for which combined datasets were avail-

able, including soil and tree species in-

formation, climatic data, and atmos-

pheric deposition (throughfall).

Part of the variation in the abundance

of the various species occurring in the

ground vegetation could be explained

by tree species, actual soil situation and

climate, mainly in terms of precipita-

tion and temperature. (Tab. 4-1). ‘Rich’

soils with high pH, high base satura-

tion and high availability of base cati-

ons, as well as southern climates and

oak forests seem to determine high

plant diversity. Th e impact of nitro-

gen deposition was lower but statis-

tically signifi cant. Deposition eff ects

may partly be hidden because of the

relationship between acid deposition

and actual soil pH on the plot, which

was an important variable explaining

ground vegetation composition. In ad-

dition, the results are only related to the

spatial distribution of species. Related

studies show that temporal changes in

ground vegetation composition can be

infl uenced by atmospheric deposition.

Th is suggests a stronger infl uence of

deposition on ground vegetation then

presented in these results. Future data

collection will allow a more appropri-

ate assessment of deposition impacts

on vegetation changes.

4.3 Single species in relation to en-

vironmental factors

Relationships between the occurrence

probability of individual species and en-

vironmental factors were investigated

for 332 diff erent species. Th is was done

by relating the species occurrence to

more than 10 000 possible combina-

tions of measured Level II data. Also

these results show a predominant in-

fl uence of soil chemistry, in particular

pH, on the occurrence of single species

and confi rm the above presented fi nd-

ings. An example for 36 selected spe-

cies against soil pH is given in Figure

4-2. Results show that most species oc-

cur on alkaline conditions whereas on

acid sites only a few specially adapt-

ed species will predominate. Th is is in

line with current views which accept

acidifi cation as a factor that negatively

aff ects biodiversity. Models are to be

developed in the coming years based

on the evaluations presented. Th ey will

allow simulations that predict chang-

es in ground vegetation composition

under changing environmental con-

ditions.

4.4 Contribution of the monitoring

programme to forest biodiversity

assessments - a future focus

Recently, ICP Forests has amended its

mandate to include contributions to

biodiversity assessments in forests by

means of the monitoring activities. In

collaboration with the European Com-

mission, a Biodiversity Working Group

has been formed to address the issue of

forest biodiversity within the pan-Eu-

ropean Monitoring Programme and to

establish links to other processes relat-

ed to biodiversity such as e.g. the Min-

isterial Conference for the Protection

of Forests in Europe (MCPFE), the Eu-

ropean Environmental Agency (EEA)

and the Convention on Biological Di-

versity (CBD).

Th e programme’s Working Group on

Biodiversity has revised the wealth of

approaches and methodologies pres-

ently available in the fi eld of forest bi-

odiversity assessment. As a contribu-

tion of the programme, the group has

suggested to adopt the stand-scale

structural approach to characterise

plant diversity in forests. Important

consideration is also given to its re-

lationships to measured environmen-

tal factors.

Th e stand-scale structural approach

uses the description of the forests

stand as an indicator of forest biodi-

versity. Th e assumption behind this is

that the more structurally diverse a for-

est stand is (e.g. in terms of the presence

or absence of vertical and horizontal

layers), the greater the range of habi-

tat types that may be associated with

that stand, thus suggesting a greater bi-

odiversity potential. Necessary data for

structural descriptions including forest

growth, stand age, number of tree spe-

cies and remote sensing already exist in

the present database. Others like stand

history, canopy closure and manage-

ment regime might be subject of addi-

tional assessments. Th ese parameters

may be related to the occurrence and

distribution of plant communities re-

corded at the plots. Th e importance of

forest deadwood to biodiversity is now

widely recognised and a measurement

of this too may be added as an assess-

ment parameter.

Th e Task Force of ICP Forests agreed

to conduct a test phase aiming to spec-

ify the possible contributions of the

programme in the fi eld of forest bio-

diversity assessments. As a fi rst step,

the possible use of existing data that

may contribute to the issue of biodver-

sity in forests will be examined at na-

tional and European level.

F :

UNECE EC, ; D V, W., G.J.

R, H. D, D. Z, D.

A, P. N, M. P, J. A,

J.C.H. V E. V. I

M F E

E. T R . UNECE

EC, G B, .

Variable group Explained variance

Actual soil situation 7.6%

Temperature, precipitation 5.6%

Tree species 4.1%

Deposition 3.3%

Total 20.6%

Table 4-1: Percentage explained var-

iance of the species abundances that

could be ascribed to the four main

groups of variables based on 194

plots.

Species that prevail at acid soils (low

pH) are: Deschampsia fl exuosa, Cal-

luna vulgaris, Calamagrostis villo-

sa, Vaccinium myrtillus, Vaccinium

vitis-idaea, Picea abies, Sorbus au-

cuparia.

Species that prevail at intermediate

soils are: Galium odoratum, Mel-

ica unifl ora, Anemone nemorosa,

Veronica offi cinalis, Hedera helix,

Carex sylvatica.

Species that prevail at alkaline soils

(high pH) are: Ajuga reptans, Viola

alba, Melittis melissophyllum, Dac-

tylis glomerata, Sorbus domestica,

Cardamine bulbifera, Silene italica,

Digitalis lutea, Festuca heterophyl-

la, Daphne laureola, Cruciata gla-

bra, Ruscus aculeatus, Carex fl acca,

Stachys offi cinalis, Rubus caesius, Poa

nemoralis, Carpinus betulus, Mercu-

rialis perennis, Solidago virgaurea,

Rosa arvensis, Luzula forsteri, Ru-

bus idaeus, Prunus spinosa, Rubus

ulmifolius, Arum maculatum.

Pro

bab

ilit

y o

f o

ccu

rren

ce

. C

Approximately a third of Europe is

covered by forests. Th ese extensive

ecosystems are partly aff ected by the

deposition of atmospheric pollutants.

Th ese inputs act within a complex of

other anthropogenic and natural stress

factors.

Th e monitoring programme of EU

and ICP Forests joins experts from 39

countries. It operates nearly 7 000 plots

throughout Europe and maintains ef-

fective communication infrastructures

to policy players and the public. It has

become an essential source of infor-

mation in the fi elds of clean air policy

and atmospheric pollution also taking

into account their relations to sustain-

able forest management, biodiversity

and climate change.

Time trends of its large-scale data on

forest condition show an overall dete-

rioration in crown condition again over

the past fi ve years, although the level of

damage is lower compared to the peak

in the mid-1990ies. More than 20% of

all trees assessed in 2001 were classi-

fi ed as damaged. For the fi rst time cor-

relations between deposition and dete-

riorating crown condition of the trees

were clearly shown in large-scale eval-

uations based on 1 300 plots of pine

trees and nearly 400 beech plots. Fur-

thermore, insect and fungi attacks and

unfavourable weather conditions have

had an impact on forest condition.

Under the Intensive Monitoring Pro-

gramme, total deposition has been cal-

culated for more than 200 plots. Inputs

of nitrogen from 1995 to 1999 mostly

range between 3.5 and 39 kg per hec-

tare and year with an average value

of 19 kg. Average sulphur inputs are

around 12.5 kg and range mostly be-

tween 3 and 29 kg. Th e eff ects of these

depositions depend on the sensitivity

of the ecosystems. Critical loads for

nitrogen and acidity have been calcu-

lated which express the highest quanti-

ty of inputs tolerable for specifi c plots.

Results show that the forests in Scandi-

navia are particularly sensitive. Critical

loads for nitrogen and acidity were ex-

ceeded by present depositions on large

parts of the plots. Based on the offi cial

UNECE manual for calculation of crit-

ical loads, the plotwise results largely

based on measurement data are im-

portant cornerstones for the partner

programme of ICP on Modelling and

Mapping, which produces area relat-

ed maps for Europe largely based on

estimated data.

Th e UNCED conference in Rio de Ja-

neiro in 1992 expressed a serious con-

cern about the world-wide loss of bio-

diversity and considered atmospheric

deposition as one of the factors that

might be responsible for this. Th e

ground vegetation data of the mon-

itoring programme in relation to the

measured environmental infl uences

now shows that the present acidity

status of the soil is clearly related to

the species occurrence. Impacts of ni-

trogen deposition were found for some

species. Additional important environ-

mental infl uences were precipitation,

temperature and the tree species grow-

ing on the plots. Th e programme has

recognised the importance of the bi-

odiversity issues and a newly estab-

lished working group is now respon-

sible for intensifi ed assessments and

evaluations that might in the future

make it possible to quantify environ-

mental impacts on fl oristic biodiver-

sity in forests.

In the 16 years of its existence the

forest monitoring programme of ICP

Forests and EU has been eff ective as

a promoter, supporter and creator of

awareness in the scientifi c, political

and public areas. Its growing datasets

and its infrastructure have become in-

creasingly interesting for other organ-

isations and projects and at the same

time the widened scope of activities

require competent partners. In partic-

ular in the Nordic countries the pro-

gramme’s monitoring data are linked

to the national forest inventories. Also,

their use for monitoring Natura 2000

habitat types is under discussion. Th e

work of ICP Forests and EU takes into

account international processes like the

Convention on Biodiversity (CBD) and

the Framework convention on Climate

Change (FCCC) and benefi ts, for exam-

ple, from co-operation with the Minis-

terial Conference on the Protection of

Forests in Europe (MCPFE) and with

deposition monitoring networks in oth-

er parts of the world.

Beech Forest, Germany .

Participating

countries

Forest

area

(1000 ha)

% of

forest area

Grid

size

(km x km)

No. of

sample

plots

No. of

sample

trees

Defoliation o all species

by class (aggregates),

national surveys

0 1 2-4

Albania 1028 35.8 10 x 10 216 6480 51.5 38.3 10.2

Austria 3878 46.2 8.7 x 8.7 260 7002 57.7 32.6 9.7

Belarus 6001 28.9 16 x 16 407 9652 18.0 61.3 20.7

Belgium 691 22.8 4² / 8² 142 3374 42.1 40.0 17.9

Bulgaria 3314 29.9 4²/8²/16² 120 4323 31.6 34.6 33.8

Croatia 2061 36.5 16 x 16 81 1941 36.1 38.9 25.0

Cyprus 298 32.2 16x16 15 360 25.8 65.3 8.9

Czech Republic 2630 33.4 8²/16² 139 6808 11.3 36.6 52.1

Denmark 445 10.3 7²/16² 52 1248 58.6 34.0 7.4

Estonia 2249 45.7 16 x 16 89 2136 49.0 42.5 8.5

Finland 20032 65.8 16² / 24x32 454 8579 56.8 32.3 10.9

France 14591 26.6 16 x 16 519 10373 44.2 35.5 20.3

Germany 10264 28.9 16² / 4² 446 13478 35.7 42.4 21.9

Greece a) 2512 19.5 16 x 16 76 1792 38.8 39.5 21.7

Hungary 1787 19.2 4 x 4 1141 26808 37.0 41.8 21.2

Ireland 436 6.3 16 x 16 21 420 55.2 27.4 17.4

Italy 8675 28.8 16 x 16 265 7351 20.3 41.3 38.4

Latvia 2888 44.7 8 x 8 365 8695 18.2 66.2 15.6

Liechtenstein 8 50.0 no survey in 2001

Lithuania 1858 28.5 8x8/16x16 286 6664 14.6 73.7 11.7

Luxembourg 89 34.4 no survey in 2001

Rep. of Moldova 318 9.4 2 x 2 580 14058 32.4 30.7 36.9

Th e Netherlands 334 9.6 16 x 16 11 231 56.3 23.8 19.9

Norway 12000 37.1 9²/18² 1647 7891 32.0 40.8 27.2

Poland 8756 28.0 16 x 16 1180 23600 9.9 59.5 30.6

Portugal 3234 36.4 16 x 16 144 4320 46.3 43.6 10.1

Romania 6244 26.3 4 x 4 4221 110190 62.5 24.2 13.3

Russian Fed. b) 7610 72.2 varying 130 2966 41.7 48.5 9.8

Slovak Republic 1961 40.0 16 x 16 110 4241 15.5 52.8 31.7

Slovenia 1099 54.2 16 x 16 41 984 31.4 39.7 28.9

Spain 11792 23.4 16 x 16 620 14880 28.9 58.1 13.0

Sweden 23400 57.1 varying 4139 16442 52.1 30.4 17.5

Switzerland 1186 28.7 16 x 16 49 1073 33.7 48.1 18.2

Turkey 20199 25.9 no survey in 2001

Ukraine 9316 15.4 16 x 16 71 1685 6.1 54.3 39.6

United Kingdom 2156 8.9 random 341 8184 32.4 46.5 21.1

Yugoslavia 2858 2.8 16 x 16 114 2674 65.2 20.8 14.0

TOTAL 198198 26.7 varying 18492 340903

Annex I: Forests, surveys and defoliation classes in European countries (2001)

– Results of national surveys as submitted by National Focal Centres -

Annex II: Defoliation of all species (1990-2001)

– Results of national surveys as submitted by National Focal Centres –

a) Excluding maquis. b) Leningrad and Pskov regions.

Note that some diff erences in the level of damage across national borders may be at least partly due to diff erences in

standards used. Th is restriction, however, does not aff ect the reliability of the trends over time.

a) Only trees older than 60 years assessed until 1997; only conifers assessed in 1990 b) Due to methodological changes, only

the time series 1990-94 and 1997-2001 are consistent, but not comparable to each other. c) For 1990, only data for former

Federal Republic of Germany. d) Excluding maquis. e) Due to methodological changes, only the time series 1989-1996 and

1997-2001 are consistent, but not comparable to each other f ) Only Kaliningrad and Leningrad Regions. g) Th e diff erence

between 1992 and subsequent years is mainly due to a change of assessment method in line with that used in other States.

Note that some diff erences in the level of damage across national borders may be at least partly due to diff erences in

standards used. Th is restriction, however, does not aff ect the reliability of the trends over time.

Participating

countries

All species change

%

pointsDefoliation classes 2-4

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2000/

2001

Albania 9.8 9.9 10.1 10.2 0.1

Austria 9.1 7.5 6.9 8.2 7.8 6.6 7.9 7.1 6.7 6.8 8.9 9.7 0.8

Belarus 54.0 29.2 29.3 37.4 38.3 39.7 36.3 30.5 26.0 24.0 20.7 -3.3

Belgium 16.2 17.9 16.9 14.8 16.9 24.5 21.2 17.4 17.0 17.7 19.0 17.9 -1.1

Bulgaria 29.1 21.8 23.1 23.2 28.9 38.0 39.2 49.6 60.2 44.2 46.3 33.8 -12.5

Croatia 15.6 19.2 28.8 39.8 30.1 33.1 25.6 23.1 23.4 25.0 1.6

Cyprus 8.9

Czech Rep. a) 45.3 56.1 51.8 57.7 58.5 71.9 68.6 48.8 50.4 51.7 52.1 0.4

Denmark 21.2 29.9 25.9 33.4 36.5 36.6 28.0 20.7 22.0 13.2 11.0 7.4 -3.6

Estonia only conifers assessed 8.7 8.7 7.4 8.5 1.1

Finland 17.3 16.0 14.5 15.2 13.0 13.3 13.2 12.2 11.8 11.4 11.6 11.0 -0.6

France b) 7.3 7.1 8.0 8.3 8.4 12.5 17.8 25.2 23.3 19.7 18.3 20.3 2.0

Germany c) 15.9 25.2 26.4 24.2 24.4 22.1 20.3 19.8 21.0 21.7 23.0 21.9 -1.1

Greece d) 17.5 16.9 18.1 21.2 23.2 25.1 23.9 23.7 21.7 16.6 18.2 21.7 3.5

Hungary 21.7 19.6 21.5 21.0 21.7 20.0 19.2 19.4 19.0 18.2 20.8 21.2 0.4

Ireland 5.4 15.0 15.7 29.6 19.7 26.3 13.0 13.6 16.1 13.0 14.6 17.4 2.8

Italy e) 16.3 16.4 18.2 17.6 19.5 18.9 29.9 35.8 35.9 35.3 34.4 38.4 4.0

Latvia 36.0 37.0 35.0 30.0 20.0 21.2 19.2 16.6 18.9 20.7 15.6 -5.1

Liechtenstein 16.0

Lithuania 20.4 23.9 17.5 27.4 25.4 24.9 12.6 14.5 15.7 11.6 13.9 11.7 -2.2

Luxembourg 20.8 20.4 23.8 34.8 38.3 37.5 29.9 25.3 23.4

Rep. of Moldova 50.8 40.4 41.2 29.1 36.9 7.8

Th e Netherlands 17.8 17.2 33.4 25.0 19.4 32.0 34.1 34.6 31.0 21.8 19.9 -1.9

Norway 17.2 19.7 26.2 24.9 27.5 28.8 29.4 30.7 30.6 28.6 24.3 27.2 2.9

Poland 38.4 45.0 48.8 50.0 54.9 52.6 39.7 36.6 34.6 30.6 32.0 30.6 -1.4

Portugal 30.7 29.6 22.5 7.3 5.7 9.1 7.3 8.3 10.2 11.1 10.3 10.1 -0.2

Romania 9.7 16.7 20.5 21.2 21.2 16.9 15.6 12.3 12.7 14.3 13.3 -1.0

Russian Fed. f) 10.7 12.5 9.8

Slovak Rep. 41.5 28.5 36.0 37.6 41.8 42.6 34.0 31.0 32.5 27.8 23.5 31.7 8.2

Slovenia 18.2 15.9 19.0 16.0 24.7 19.0 25.7 27.6 29.1 24.8 28.9 4.1

Spain 4.7 7.4 12.3 13.0 19.4 23.5 19.4 13.7 13.6 12.9 13.8 13.0 -0.8

Sweden only conifers assessed 14.2 17.4 14.9 14.2 13.2 13.7 17.5 3.8

Switzerland 15.5 16.1 12.8 15.4 18.2 24.6 20.8 16.9 19.1 19.0 29.4 18.2 -11.2

Turkey

Ukraine 2.9 6.4 16.3 21.5 32.4 29.6 46.0 31.4 51.5 56.2 60.7 39.6 -21.1

United Kingd. g) 39.0 56.7 58.3 16.9 13.9 13.6 14.3 19.0 21.1 21.4 21.6 21.1 -0.5

Yugoslavia 9.8 3.6 7.7 8.4 11.2 8.4 14.0 5.6

Annex III

Common beech Fagus sylvatica

Corsican pine Pinus nigra var. maritima

European oak Quercus robur

Holm oak Quercus ilex

Maritime pine Pinus pinaster

Norway spruce Picea abies

Red oak Quercus rubra

Scots pine Pinus sylvestris

Sessile oak Quercus petraea

Annex IV

Photo, page Author

title top left, bottom right: R. Fischer

top right: W. Seidling

bottom left: P. Roskams

6 E. Oksanen

7, 12, 26, 27 top R. Fischer

9, 27 bottom Institute for World Forestry,

Hamburg, Germany

10, 28 left W. Seidling

17 E. Ulrich

18, 19 General Directorate of Nature

Conservation, Madrid, Spain

19, big photo Antonio Moreno. CENEAM.

Ministerio de Medio Ambiente,

Madrid

20, 22 P. Roskams

28 middle and right A. Reif

31 R. Schönemund

Participating countries:

Albania Lithuania Portugal

Austria Luxembourg Romania

Belarus Republic of Moldova Russian Federation

Belgium Th e Netherlands Slovak Republic

Bulgaria Norway Slovenia

Canada Poland Spain

Croatia Germany Sweden

Cyprus Greece Switzerland

Czech Republic Hungary Turkey

Denmark Ireland Ukraine

Estonia Italy United Kingdom

Finland Latvia United States

France Liechtenstein Yugoslavia

For further information please contact:

Federal Research Centre for Forestry and Forest Products

PCC of ICP Forests

attention Dr. M. Lorenz, R. Fischer

Leuschnerstr. 91

D-21031 HAMBURG

European Commission

DG AGRI, Fl. 3

Rue de la Loi 130

B – 1040 Brussels

F :

http://www.icp-forests.org

(ICP Forests)

http://europa.eu.int/comm/agriculture

(European Commission)

http://www.fi mci.nl

(Forest Intensive Monitoring Co-ordinating Institute)

A F, D-F

M R, D-W

ISSN 1020-587X

Printed in Germany